Expedite Biomarker Discovery with Human-Centric AI

Biomarkers are pivotal throughout drug development, from discovery to market, playing key roles in unravelling drug mechanisms, providing prognostic insights and assessing treatment efficacy. In fact, in recent years the FDA have approved drugs earlier (including after just Phase I trials) in cases where biomarker selection enabled remarkable response rates.¹ This underscores the transformative impact of biomarkers in accelerating the drug development pipeline.

Despite the clinical promise, biomarker development is challenging. There are substantial obstacles, from disease heterogeneity and rigorous validation requirements to the inability to extract meaningful biomarker insights from extensive biological data.² With over 1M+ biomarkers on PubMed and a 8-9% annual growth in research papers,³ identifying biomarkers with suitable sensitivity and specificity is time-consuming and subject to selection bias (Figure 1). This provides limited opportunity for discovering hidden knowledge and generating hypotheses, leading to missed opportunities.

With the cost of bringing a drug to market estimated at $1-2 billion,⁴ there is increasing pressure for pharmaceutical companies to optimize their R&D strategies, streamline clinical trials and improve success rates. Using Causaly’s human-centric AI, scientists can accelerate the discovery of the most relevant biomarkers from the entire volume of biomedical literature in seconds, saving time, reducing bias and making better decisions. Crucially, Causaly’s AI gives a view of all scientific evidence with full transparency, allowing the user to scrutinize and exercise their own judgement on the presented evidence.

In our latest Spotlight On Biomarkers Report, we delve into biomarkers of disease progression, treatment response, drug resistance and survival in neurological, oncological and neuromuscular indications. Figure 2 highlights some key insights.

The utility of biomarkers in drug development cannot be understated, yet biomarker discovery is fraught with challenges, including disease heterogeneity and the difficulty in deriving insights from vast biological data. With an average of 2 papers per minute added to PubMed,³ identifying relevant biomarkers is time-consuming and prone to bias. Human-centric AI can manage the data overload by machine-reading the volume of biomedical literature and extracting only the most relevant biomarker insights. Importantly, Causaly provides full transparency empowering scientists to exercise their own judgement, thus fostering  informed decision-making.